21

Statistical Expectations: What’s Your Size?

THE DISTINCTIVE FACT about statistics in the United States was not their rise as a learned specialty. One by-product of democracy was an unprecedented popular diffusion of statistics, and in the twentieth century a new kind of number consciousness captured the public mind. The ballot box and the census were only two of many American ways of asking, How many? How much? How big?

In developing and reinforcing this number consciousness, nothing was more important than American industry and the American Standard of Living. Increasing pressures to make all sorts of commodities in large factories for the use of everybody led Americans to apply the techniques of statistics to everyday life in myriad ways.

A COMMONPLACE EXAMPLE was in the new democracy of clothing. We have already seen how the rise of a ready-made-clothing industry moved the center of garment making in the United States from the home to the factory. Among the many circumstances that had made this possible was the growth of a new branch of statistical science. “Anthropometry” (a word which entered English just before the mid-nineteenth century) was “the measurement of the dimensions of man” with a view to discovering those which most commonly recurred and to defining the differing proportions of the parts of the body in different ages, classes, and races. Until the mid-nineteenth century it was generally assumed, on the basis of common sense strengthened by religious notions of the uniqueness of each individual soul, that the dimensions of every man’s body were quite different from those of every other man. If that was the case, then it was futile to try to manufacture large quantities of clothing that would actually fit different wearers. If a man wanted a proper fit he would have to make his own clothing or employ a tailor.

The Civil War, with its sudden demand for uniforms in great quantity, produced a fund of new information on the common dimensions of the human body. And this disclosed the crucial fact that certain sets of measurements tended to recur with predictable regularity. If a man’s waist measured 38 inches and his sleeve-length was 34 inches, then in eight cases out of ten his shoulders would have a certain breadth. This simple discovery made it possible to manufacture well-fitting garments for a large population. Simple though it was, the discovery was essential to the ready-made-clothing industry, for without it no shop could be provided with a disposable stock which would fit the customers. At the end of the war, these measurements were available to guide clothing manufacturers in their efforts to supply ready-made civilian garments to an unprecedented market: hundreds of thousands of demobilized soldiers; a flood of new immigrants; prospering inhabitants of mushrooming cities like Chicago, Omaha, and Denver; and workmen in thousands of new factories. Now for the first time middle-class Americans, too, could find store-boughten clothing that was attractive and well-fitting.

Essential to all this was the increase and diffusion of information about bodily measurements. In 1880 appeared Daniel Edward Ryan’s epoch-making book Human Proportions in Growth: Being the Complete Measurements of the Human Body for Every Age and Size during the Years of Juvenile Growth. Now at last Ryan offered a scientific guide for standardizing measurements in men’s, boys’, and juveniles’ ready-made clothing. Ryan’s system included practical charts for deriving the other bodily proportions from his statistics on body circumference and height. By 1889 Professor Franz Boas was teaching courses in anthropometry and anthropological statistics at Clark University. And these statistics on bodily growth, useful in countless other ways (for example in better design for school furniture), incidentally provided the raw material for a flourishing new science of educational psychology.

By the early twentieth century, then, for the first time in history, it was possible for a man to walk into a clothing store, indicate that he was a “42” and put on a jacket that with little or no alteration would satisfy a fastidious eye. People thus began to think of themselves as belonging to certain “sizes”—in shoes, shirts, trousers, and hats—another sign that Americans had joined statistical communities.

TO MAKE STANDARDIZED PRODUCTS that would meet the customers’ expectations, it was necessary to have standardized units of measurement. We have become so accustomed to knowing what we mean by a minute, a foot, or a gallon that we forget how recently these simple terms were given clear definition. “The system of weights and measures in customary use,” a leading physicist observed in 1887, “is so confusing, so unscientific, and, in some instances, apparently so contradictory that it is difficult to write of it, even briefly, without falling into error.” As late as 1892 there were eight different “authoritative” values for the U. S. gallon.

These “laissez-faire standards” left the way open for any manufacturer to use whatever definition best served his advertising needs at the moment. In new industries this was especially troublesome. The president of the American Institute of Electrical Engineers complained in 1900 that because the English standard of light-measurement (which had just been adopted by the Office of Weights and Measures in the United States Coast and Geodetic Survey) was so indefinite, American scientific laboratories were using the more precise German standard. At the same time, the electric-light industry, dissatisfied with both the English and the German standards, adopted a standard of its own. As a result an electric lamp requiring 10 amperes at a pressure of 45 volts was being described in the advertising as “2,000 candlepower,” when the same lamp (measured by British or German standards) actually would have been rated at only 400 to 500 candlepower.

Amidst this confusion, the construction industry found it difficult to figure costs or to meet schedules. An expert writing in Scientific American in 1896 reported that a prudent builder had to order about 20 percent more material than he actually needed for the job, in order to allow for uneven quality and crude standards. By 1900, when Americans had become adept at manufacturing all sorts of industrial measuring devices, it was still common to send these devices abroad, to England or more usually to Germany, to be calibrated.

Among the obvious sources of confusion were the extent and variety of the United States, the fast growth of industry, and the rapid rate of change. The federal system of government, and especially the notion that police powers and day-to-day regulation were the sacred province of the separate states, perpetuated confusion. But the Constitution (art. I, sec. 8) had given to the Congress (along with the power “to collect Taxes, Duties, Imposts and Excises” and “to coin Money, regulate the Value thereof, and of foreign Coin”) the power to “fix the Standard of Weights and Measures.” The establishment of nationally uniform standards of weights and measures, when it finally came, had to be justified indirectly (in deference to states’ rights) as a way of satisfying the constitutional requirements concerning coinage and the requirement that “all Duties, Imposts and Excises shall be uniform throughout the United States.”

For more than a century, Congress struggled with various devices for establishing a sufficient uniformity of weights and measures which would yet not infringe on the powers of the states or the independence of businessmen. Jefferson, Madison, and John Quincy Adams all made proposals, but the Congress, fearing excessive federal control, could not agree. In 1828, Congress managed to adopt the British troy pound of 1758 as the standard for American coinage. Albert Gallatin, a former Secretary of the Treasury, had, when he was minister to Great Britain in 1827, secured a brass copy of the old British troy pound which was deposited with the Director of the Mint in Philadelphia. Then copies were supplied to all the mints as the basis for the weight of a pound of gold. But when the Houses of Parliament burned in 1834, the British prototype was destroyed, and thereafter the brass pound at Philadelphia had to be defined indirectly by reference to the British avoirdupois pound. In any case, as Charles S. Peirce (the versatile and erratic philosopher who was destined to become the chief metaphysician of American pragmatism) observed in 1884 when he was head of the Office of Weights and Measures, the troy pound was quite unsuitable for precision weights. He explained that since the standard troy pound had never been weighed in a vacuum to determine its true weight, and the United States government had no apparatus capable of weighing in a vacuum, the weight of the American pound was “not known.” Nevertheless, for lack of anything better, this indeterminate “pound” remained the American standard for coinage until 1911.

A federal office for weights and measures had appeared within the Department of the Treasury in response to complaints about lack of uniformity in the customs collections at different ports. The first head of the Coast Survey, Ferdinand Rudolph Hassler, a recent Swiss immigrant, found that there were no two custom houses which used the same measure for the pound or the bushel. The practical result, of course, was that customs duties were not uniform. After about 1830, Hassler made strenuous efforts to establish uniform standards, but congressional fears of federal domination forced Hassler to try to accomplish his purpose by persuading each of the states separately to adopt his proposals. This was less a scientific than a political enterprise, and his progress was slow.

Then, after the Spanish-American War of 1898, the United States plunged into world politics. In the preceding decade American exports of manufactures had nearly doubled. When the nation was a leader in world trade it was hard any longer to postpone establishing uniform American standards of measurement. Customers all over the world wanted assurance of the quality of American products. The leader of a new campaign for a national standardizing laboratory was Lyman J. Gage, a self-made Chicago businessman who had headed the board of the World’s Columbian Exposition in 1893, and then as President William McKinley’s Secretary of Treasury had financed the Spanish-American War. In 1900 Gage urged on Congress that the nation could no longer lean on others for the definition of its standards; American manufacturers needed a national effort to “secure the requisite degree of uniformity and accuracy” in their products. Congress enacted Secretary Gage’s proposal in 1901, incidentally changing the name from Gage’s more explicit “National Standardizing Bureau” to the “National Bureau of Standards.” The total appropriation for 1900 for the old Office of Weights and Measures had been $9,410. In congressional debate on the bill, some congressmen had objected that the head of the new Bureau was scheduled to receive $6,000 a year, almost as high a salary as that of the Secretary of the Treasury himself. Secretary Gage knowingly replied that while “almost anybody will do for the Secretary of the Treasury … it takes a very high-grade man to be chief of a bureau like this.”

The “high-grade man” who came to Washington to create a National Bureau of Standards fulfilled Gage’s expectations. Samuel Wesley Stratton, son of an Illinois farmer, had worked his way through the University of Illinois, where he established a novel course in electrical engineering, and then he had gone on to the newly organized University of Chicago in 1892. There he worked with Albert Michelson, experimenting on the speed of light, until he joined the Navy during the Spanish-American War. At Gage’s request Stratton, almost single-handed, had drawn the bill for the new Bureau. Then, one witness observed, he “mesmerized” the congressional committee with his explanations and demonstrations. As first head of the Bureau, Stratton built up its equipment and enlisted a brilliant scientific staff with a passion for research and experiment. Within twenty years he had given the Bureau a leading role in American industrial life: the transfer of the Bureau from the Department of the Treasury to the Department of Commerce in 1903 emphasized its dynamic new function.

When Stratton left the Bureau in 1923, the United States had become a world leader in techniques of testing, measuring, and standardizing. The Bureau was taking the initiative in testing all sorts of materials; it established standards for the new world of electricity, including the vast and controversial realm of public utilities, and it developed new precision techniques for measuring light and gas. The Bureau’s brochures educated the consumer public on the qualities of household materials and appliances. “Usually, books on radio communication are fairly bristling with mathematics,” Thomas A. Edison observed in 1919, “and I am at a loss trying to read them,” but the Bureau’s publication on radio was “the greatest book on this subject” he had ever read. During World War I, the Bureau experimented with substitutes for leather, paper, and textiles, and helped develop ships made of concrete. Wartime needs aided the Bureau’s effort to persuade the nation’s manufacturers to standardize, and after the war, the Bureau led a “Crusade for Standardization” as part of Secretary of Commerce Herbert Hoover’s campaign against waste.

In the following decades and in each succeeding national crisis the Bureau of Standards played a leading, and usually a novel, role. In the Depression years the Bureau reviewed the NRA “codes” of fair competition and consumer standards. In World War II it sought ways to conserve petroleum, and joined the effort to make synthetic rubber and other substitutes for scarce materials. In the quest for nuclear power, and in the building of the atomic bomb, the Bureau provided essential research on uranium, developed a new type of fuse, and became integral to the success of the whole enormous enterprise. In the exploration of space, too, the Bureau helped in scores of ways, developing and refining instruments for testing and measuring in uncharted new ranges of dimension and in unprecedented extremes of temperature.

By 1970 the National Bureau of Standards, just outside the capital, had become a 70-acre shrine to the nation’s concern with quantity, and its uses of quantity to define quality. There the high priests of measurement reinforced the nation’s faith in the language of numbers.

If you find an error please notify us in the comments. Thank you!